Instance Object Identification on COST (val)
73.3CSVCoder LLaVA-1.5-13b
Evaluation results
| Method | Links | ||
|---|---|---|---|
| VCoder LLaVA-1.5-13bInput Tokens=(seg) + (img) + (query)2023.12 | 73.3 | 25 | |
| VCoder LLaVA-1.5-7bInput Tokens=(seg) + (img) + (query)2023.12 | 71.1 | 26.9 | |
| COST IT LLaVA-1.5-7bInput Tokens=(img) + (query)2023.12 | 67.5 | 30.3 | |
| ImCoder LLaVA-1.5-7bInput Tokens=(img) + (img) + (query)2023.12 | 64 | 29.4 | |
| LLaVA-1.5-7bInput Tokens=(img) + (query)2023.12 | 50.3 | 75.9 | |
| LLaVA-1.5-13bInput Tokens=(img) + (query)2023.12 | 49.9 | 75 | |
| CogVLM-17bInput Tokens=(img) + (query)2023.12 | 43.5 | 86.2 | |
| InstructBLIP Vicuna-7bInput Tokens=(img) + (query)2023.12 | 25.3 | 91.9 | |
| Soft-Prompted LLaVA-1.5-7bInput Tokens=(prompt) + (img) + (query)2023.12 | 18.4 | 72.2 | |
| MiniGPT-4 LLaMA-2-7bInput Tokens=(img) + (query)2023.12 | 5.6 | 97.7 |